Deep learning PyTorch library for time series forecasting, classification, and anomaly detection (originally for flood forecasting).

Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting, AAAI 2019, pytorch version

Some TrafficFlowForecasting Solutions(交通流量预测解决方案)

Code for Deep-learning Architecture for Short-term Passenger Flow Forecasting in Urban Rail Transit

ST-SSL (STSSL): Spatio-Temporal Self-Supervised Learning for Traffic Flow Forecasting/Prediction

DSTAGNN: Dynamic Spatial-Temporal Aware Graph Neural Network for Traffic Flow Forecasting, which is accepted at ICML2022.

paper : <Spatial-Temporal Transformer Networks for Traffic Flow Forecasting>

No description available

The official code for "Spatial temporal transformer networks for traffic flow forecasting"

[CVPR 2025] MoFlow: One-Step Flow Matching for Human Trajectory Forecasting via Implicit Maximum Likelihood Estimation Distillation

Spatiotemporal Adaptive Gated Graph Convolution Network for Urban Traffic Flow Forecasting

Public tutorials of using Flow Forecast for forecasting and classifying time series data

Code for Multi-graph convolutional network for short-term passenger flow forecasting in urban rail transit

An offical implementation of STDN: "Spatiotemporal-aware Trend-Seasonality Decomposition Network for Traffic Flow Forecasting." (AAAI 2025)

KDD'2025. This project aims to explore the application of hypergraph spatio-temporal learning and Large Language Models (LLMs) in traffic flow demand forecasting

地铁乘客流量预测

a model for traffic flow forecast

Spatial-Temporal Transformer Networks for Traffic Flow Forecasting

Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting, ICLR 2025

A correlation information-based spatiotemporal network for traffic flow forecasting

[IJCAI'2022] FOGS: First-Order Gradient Supervision with Learning-based Graph for Traffic Flow Forecasting

This repo contains a PyTorch-based AE-ConvLSTM model for fluid flow prediction. It can forecast 5–10 time steps per forward pass and over 100 steps in rollout. The model is trained using both data-driven and physics-constrained methods. While Burgers' equation was learned successfully with PDE loss, unsteady Navier–Stokes training did not converge.

CNN-Bidirectional LSTM network to forecast long term traffic flow in Madrid.

Graph Neural Networks utilization for Spatiotemporal graphs. These methods will be applied into the problem of forecasting traffic flow on PEMS-Bay, METR-LA and Seattle Loop Datasets

ASTGCN, LSTM, GCN LSTM with fused result from previous week, day, hour

Paper STGAT: Spatial-Temporal Graph Attention Networks for Traffic Flow Forecasting

Article on using deep learning to extract order flow information from the limit order book and forecast directional moves

Code for paper "Continuous and Distribution-free Probabilistic Wind Power Forecasting: A Conditional Normalizing Flow Approach" https://arxiv.org/abs/2206.02433

Develop a deep learning model capable of predicting traffic flow in urban environments. The model will utilize historical traffic data, weather conditions, and road configurations to forecast traffic patterns. This information can be invaluable for traffic management systems, helping to optimize traffic signals and reduce congestion, ultimately.

[ICDE 2023] Dynamic hypergraph structure learning for traffic flow forecasting